Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Audio-Visual Speech Recognition on LRS3 Noisy w.r.t Viewpoint 5° Offset (test)
Loading...
3.86
WER
M2S-AVSR
3.444
6.252
9.06
11.868
Jun 4, 2026
WER
Updated 1mo ago
Evaluation Results
Method
Method
Links
WER
M2S-AVSR
Modality=A+V, Labeled...
2026.06
3.86
MMS-Llama
Modality=A+V, Labeled...
2026.06
3.95
M2S-AVSR
Modality=A+V, Labeled...
2026.06
4
Auto-AVSR
Modality=A+V, Labeled...
2026.06
4.63
M2S-AVSR
Modality=A+V, Labeled...
2026.06
4.83
MMS-Llama
Modality=A+V, Labeled...
2026.06
4.9
M2S-AVSR
Modality=A+V, Labeled...
2026.06
5.58
Llama-AVSR
Modality=A+V, Labeled...
2026.06
6
Whisper-Flamingo
Modality=A+V, Labeled...
2026.06
6.36
Whisper-Flamingo
Modality=A+V, Labeled...
2026.06
6.37
Whisper-Flamingo
Modality=A+V, Labeled...
2026.06
6.84
Whisper-Flamingo
Modality=A+V, Labeled...
2026.06
6.86
CMA
Modality=A+V, Labeled...
2026.06
7.94
AV-HuBERT
Modality=A+V, Labeled...
2026.06
9.6
AV-RelScore
Modality=A+V, Labeled...
2026.06
10.4
V-CAFE
Modality=A+V, Labeled...
2026.06
14.26
Feedback
Search any
task
Search any
task